Re: [boost] [Review Request] Multiprecision Arithmetic Library

Subject: Re: [boost] [Review Request] Multiprecision Arithmetic Library
From: Christopher Kormanyos (e_float_at_[hidden])
Date: 2012-04-06 18:35:59

<snip>
>> I've got a recursive Karatsuba template available in my catalog already.

>> Maybe we should try it out.

> For sure!

I will put the Karatsuba into my local copy and take it to the test
over the holidays. I will report on its performance as the data
become available.

<snip>
> So maybe examples are a higher priority for now?

I added a brief PDF report on generic numeric programming
and included three sample files in the SVN repo today.
https://svn.boost.org/svn/boost/sandbox/big_number/libs/multiprecision/example/generic_numerics_examples/

I can't write quickbook so I wrote LaTeX.
I can't add *.tex to SVN (my bad on some kind of mime or whatever).
I added the LaTeX text to this e-mail.

If you approve of these examples, may I ask you to translate my doc
to yours?

In my opinion, Boost.Multiprecision is a wonderful creation from
John and Paul and my preliminary contributions. There is nothing
like it in compiled code for MP numerical analysis. I have a lot
more ideas for this thing than time.

My dream list is *way* bigger than the time we have right now:

        * 1,000 digits high performance
        * Transition to competitive base-2 back-end

        * up to 1,000,000 digits for the hunters
        * PSLQ algorithm for experimental mathematics
        * Seamless interface to Python and Mathematica (in my original research paper)*But we need to go to review with what we got soon.*

<snip>

Best regards, Chris.

P.S. I also have a potential solution to my excessive guard digits.
I can show it to you with the Karatsuba stuff and, based on your approval,
commit it.